Accepted Manuscript

Diet modulates the relationship between immune gene expression and functionalimmune responses

Sheena C. Cotter, Catherine E. Reavey, Yamini Tummala, Joanna L. Randall, RobertHoldbrook, Fleur Ponton, Stephen J. Simpson, Judith A. Smith, Kenneth Wilson

PII: S0965-1748(18)30490-9

DOI: https://doi.org/10.1016/j.ibmb.2019.04.009

Reference: IB 3156

To appear in: Insect Biochemistry and Molecular Biology

Received Date: 8 January 2019

Revised Date: 1 April 2019

Accepted Date: 2 April 2019

Please cite this article as: Cotter, S.C., Reavey, C.E., Tummala, Y., Randall, J.L., Holdbrook, R., Ponton,F., Simpson, S.J., Smith, J.A., Wilson, K., Diet modulates the relationship between immune geneexpression and functional immune responses, Insect Biochemistry and Molecular Biology (2019), doi:https://doi.org/10.1016/j.ibmb.2019.04.009.

This is a PDF file of an unedited manuscript that has been accepted for publication. As a service toour customers we are providing this early version of the manuscript. The manuscript will undergocopyediting, typesetting, and review of the resulting proof before it is published in its final form. Pleasenote that during the production process errors may be discovered which could affect the content, and alllegal disclaimers that apply to the journal pertain.

MANUSCRIP

T

ACCEPTED

ACCEPTED MANUSCRIPT

PO g

ene

expr

essi

on

PO Enzyme activity Injected with X. luminescens

24 hrs

Blood sampled for gene expression and

enzyme activity Low

high

20 diets

Protein in dietProtein:carb in diet

Calo

rie c

onte

nt

MANUSCRIP

T

ACCEPTED

ACCEPTED MANUSCRIPTDiet modulates the relationship between immune gene expression and functional immune 1

responses 2

Sheena C. Cotter1, Catherine E. Reavey2, Yamini Tummala2, Joanna L. Randall2, Robert 3

Holdbrook 2, Fleur Ponton3,4, Stephen J. Simpson4, Judith A. Smith5 & Kenneth Wilson2. 4

Author affiliations: 5

1 School of Life Sciences, University of Lincoln, Brayford Pool, Lincoln LN6 7TS, UK 6

2 Lancaster Environment Centre, Lancaster University, Lancaster LA1 4YQ, UK 7

3 Charles Perkins Centre, University of Sydney, NSW 2006, Australia 8

4 Department of Biological Sciences, Macquarie University, NSW 2109, Australia 9

5 School of Forensic and Applied Sciences, University of Central Lancashire, Preston, 10

Lancashire, PR1 2HE, UK 11

Corresponding author: Sheena Cotter, School of Life Sciences, University of Lincoln, Brayford 12

Pool, Lincoln LN6 7TS, UK, Tel: +44 1522 88 6835, scotter@lincoln.ac.uk 13

14

Classification: Biological Sciences - Ecology 15

16

17

18

19

20

21

22

23

MANUSCRIP

T

ACCEPTED

ACCEPTED MANUSCRIPTAbstract 24

Nutrition is vital to health and the availability of resources has long been acknowledged as a key 25

factor in the ability to fight off parasites, as investing in the immune system is costly. Resources have 26

typically been considered as something of a “black box”, with the quantity of available food being 27

used as a proxy for resource limitation. However, food is a complex mixture of macro- and 28

micronutrients, the precise balance of which determines an animal’s fitness. Here we use a state-29

space modelling approach, the Geometric Framework for Nutrition (GFN), to assess for the first 30

time, how the balance and amount of nutrients affects an animal’s ability to mount an immune 31

response to a pathogenic infection. 32

Spodoptera littoralis caterpillars were assigned to one of 20 diets that varied in the ratio of 33

macronutrients (protein and carbohydrate) and their calorie content to cover a large region of nutrient 34

space. Caterpillars were then handled or injected with either live or dead Xenorhabdus nematophila 35

bacterial cells. The expression of nine genes (5 immune, 4 non-immune) was measured 20 h post 36

immune challenge. For two of the immune genes (PPO and Lysozyme) we also measured the 37

relevant functional immune response in the haemolymph. Gene expression and functional immune 38

responses were then mapped against nutritional intake. 39

The expression of all immune genes was up-regulated by injection with dead bacteria, but only those 40

in the IMD pathway (Moricin and Relish) were substantially up-regulated by both dead and live 41

bacterial challenge. Functional immune responses increased with the protein content of the diet but 42

the expression of immune genes was much less predictable. 43

Our results indicate that diet does play an important role in the ability of an animal to mount an 44

adequate immune response, with the availability of protein being the most important predictor of the 45

functional (physiological) immune response. Importantly, however, immune gene expression 46

responds quite differently to functional immunity and we would caution against using gene 47

expression as a proxy for immune investment, as it is unlikely to be reliable indicator of the immune 48

response, except under specific dietary conditions. 49

MANUSCRIP

T

ACCEPTED

ACCEPTED MANUSCRIPTKeywords: Nutritional ecology, host-pathogen interaction, immunity, Spodoptera, Xenorhabdus, 50

diet, bacteria, resistance, tolerance, insect, Geometric Framework 51

52

53

Introduction 54

It has long been recognised the role that “good nutrition” plays in human health, with both under-55

nutrition and obesity resulting in disease (Mokdad et al., 2001; Muller and Krawinkel, 2005; Samartin 56

and Chandra, 2001). Poor nutrition can also impact the response to parasites, with evidence for both 57

energy and protein deficits reducing the ability to fight infection (Kuvibidila et al., 1993) (Field et al., 58

2002) (Cunningham-Rundles et al., 2005). Studies have shown that starvation can compromise 59

immune capability across a broad range of host taxa. For example, laboratory mice were found to 60

have fewer T cells in the spleen and thymus during starvation, with numbers recovering once feeding 61

was reinstated (Wing et al., 1988). Furthermore, injection with Listeria monocytogenes during 62

starvation reduced the ability of the mice to develop antibodies against this bacterium (Wing et al., 63

1988). Food restriction, rather than starvation can have similar effects. Food-restricted Yellow-legged 64

gulls, Larus cachinnans, were found to have reduced cell-mediated immunity (Alonso-Alvarez and 65

Tella, 2001) and mice on a long-term calorie-restricted diet were found to die more rapidly from 66

sepsis after gut puncture than those fed ad libitum (Alonso-Alvarez and Tella, 2001). Comparable 67

responses have been shown in invertebrates; bumble bees died more rapidly during starvation if their 68

immune systems were stimulated by artificial parasites, suggesting that mounting an immune response 69

is energetically costly (Moret and Schmid-Hempel, 2000). Similarly, starved bumble bees were more 70

likely to die from a gut parasite, Crithida bombi, than hosts with adequate nutrition (Brown et al., 71

2000). 72

However, nutrition is much more complex than simply a source of energy, being a vital mixture of 73

macro- (carbohydrates, fats and proteins) and micro-nutrients (vitamins and minerals), the amount and 74

balance of which determine an animal’s fitness (Simpson et al., 2004). Several studies have examined 75

MANUSCRIP

T

ACCEPTED

ACCEPTED MANUSCRIPThow shifting the balance of macronutrients in the diet affects immune responses and the outcome of 76

infection, without restricting the availability of calories (Graham et al., 2014; Lee et al., 2006; Ponton 77

et al., 2011; Povey et al., 2009; Povey et al., 2014) . For example, caterpillars of the armyworms, 78

Spodoptera littoralis and Spodoptera exempta, show improved immune responses and markedly 79

higher survival after viral infection (Lee et al., 2006; Povey et al., 2014) and bacterial infection 80

(Povey et al., 2009) when their diet is heavily protein-biased. Furthermore, when given the 81

opportunity, infected caterpillars will “self-medicate” with protein, significantly improving their 82

chances of survival (Lee et al., 2006; Povey et al., 2009; Povey et al., 2014) . 83

The studies above strongly suggest that it is the source of the energy in the diet that is key to the 84

response to parasites, rather than the availability of energy per se. However, neither food restriction, 85

nor the manipulation of macronutrient balance alone can determine the relative importance of either 86

on host-parasite interactions. To address properly the role of nutrient availability on immunity, both 87

the balance of nutrients in the diet and their quantity need to be simultaneously manipulated. The 88

Geometric Framework for Nutrition (GFN) is a state-space model that allows the association of 89

particular nutrient intakes with outcomes of interest (Simpson and Raubenheimer, 1995), for example, 90

immunity (Ponton et al., 2011; Ponton et al., 2013). With the GFN, animals are restricted to diets in 91

which both the balance and availability of nutrients are manipulated, forcing intakes over a wide 92

region of nutrient space, encompassing both over- and under-nutrition, and thereby allowing the 93

additive and interactive effects of specific nutrients on traits of interest to be modelled (Simpson and 94

Raubenheimer, 1995). 95

The GFN approach has highlighted that the fundamental life-history trade-off between fecundity and 96

longevity is mediated by nutrients across taxa, with longevity generally peaking at low-protein, high-97

carbohydrate ratios, whilst fecundity tends to peak at much higher relative protein intakes; as such, no 98

diet can maximize both traits (Drosophila: (Lee et al., 2008); (Jensen et al., 2015); Field crickets: 99

(Maklakov et al., 2008); Queensland Fruit fly; (Fanson et al., 2009); Mice: (Solon-Biet et al., 2015)). 100

Similarly, using the GFN, it was found that different immune responses peak in different regions of 101

nutrient space, thereby indicating a nutrient-mediated trade-off within the immune system, and, as for 102

MANUSCRIP

T

ACCEPTED

ACCEPTED MANUSCRIPTfecundity and longevity, no single diet could maximize multiple immune responses (Cotter et al., 103

2011b). In a recent study, mice were restricted to one of 25 diets varying in their ratio of proteins, fats 104

and carbohydrates and energy density, and their innate immune capacity was measured. It was shown 105

that the balance of T cells indicative of healthy ageing was achieved on a low protein:NPE diet (non-106

protein energy i.e. carbohydrate plus fat), irrespective of calorie intake (Le Couteur et al., 2015). 107

However, this powerful approach has not yet been taken to assess an animal’s immune response to a 108

pathogenic challenge. 109

Insects have a comparatively simple yet effective immune system that has numerous parallels to the 110

innate immune response of vertebrates Vilmos, 1998 #1473; Leulier, 2008 #47751; Wiesner, 2010 111

#77972} . It comprises cellular and humoral components that work together to fight invading 112

pathogens. Hemocytes show phagocytic activity against microparasites, much like vertebrate 113

macrophages, and can respond to macroparasites by forming a multi-layered envelope around the 114

invader, in a process called encapsulation, which is subsequently melanised via the action of the 115

phenoloxidase (PO) enzyme (Gupta, 1991). Phenoloxidase is stored in hemocytes in the form of an 116

inactive precursor, Pro-phenoloxidase (PPO), which is activated upon detection of non-self 117

(Gonzalez-Santoyo and Cordoba-Aguilar, 2012). This recognition occurs via the detection of 118

pathogen-associated molecular patterns (PAMPs) such as the peptidoglycan or the lipopolysaccharide 119

components of fungal and bacterial cell walls. Detection stimulates either the Toll (fungi and gram-120

positive bacteria) or Imd pathways (Gram-negative bacteria), via host pattern recognition receptors 121

(PRRs) that result in the bespoke production of antimicrobial peptides and the upregulation of 122

constitutive lysozymes, which form the humoral component of the response (Ligoxygakis, 2013; 123

Wiesner and Vilcinskas, 2010). 124

The strength of the immune response can be measured using standard functional assays of 125

antimicrobial activity or PPO activity in the haemolymph, and the strength of the encapsulation 126

response or phagocytosis can be measured against synthetic parasites injected into the haemocoel (see 127

(Wilson and Cotter, 2013) and references therein). These functional responses have been shown to be 128

indicative of the ability of the animal to fight off parasites (e.g. (Lee et al., 2006; Paskewitz and 129

MANUSCRIP

T

ACCEPTED

ACCEPTED MANUSCRIPTRiehle, 1994; Povey et al., 2009) and so are arguably meaningful measures of immune investment. 130

However, gene expression is also often used as a proxy for investment in specific traits, e.g. immunity 131

(Freitak et al., 2007; Jackson et al., 2011; Woestmann et al., 2017), but few of these studies consider 132

how well the expression of the gene of interest predicts the functional response under the conditions 133

in which they are tested. 134

There has been a great deal of research examining how well gene transcripts relate to protein 135

abundance across individual genes, but with contradictory findings (Liu et al., 2016). This is not 136

surprising as there are numerous steps between gene expression and the production of the protein, all 137

of which can change the relationship between the two. In cell culture, under steady-state conditions, 138

mRNA transcripts correlate well with protein abundance, typically explaining between 40 and 80% of 139

the variation (Edfors et al., 2016; Jovanovic et al., 2015; Liu et al., 2016). However, multiple factors 140

can affect this relationship. Upregulation of gene expression in response to a perturbation is expected 141

to change the abundance of proteins concordantly, but there can be a delay in this process, such that 142

there is a time lag between mRNA levels and protein abundance, the length of which may differ 143

between genes (Gedeon and Bokes, 2012; Jovanovic et al., 2015). Some genes are constitutively 144

transcribed and translation of the protein occurs only when the correct conditions are met, known as 145

“translation on demand” (Hinnebusch and Natarajan, 2002), meaning that there is no correlation 146

between mRNA and protein levels most of the time. The majority of ecological studies consider gene 147

expression in whole animals, which are hugely more variable than cell cultures, and so we can expect 148

the relationship between gene expression and protein abundance to be further weakened in natural 149

systems. One aspect of variation in whole animals is the availability of resources. Protein production 150

is costly, consuming ~50% of the ATP in growing yeast cells (Warner, 1999), so we can expect the 151

availability of energy and amino acids to affect the speed and efficacy of translation (Liu et al., 2016). 152

This means that the relationship between the expression of a gene and its protein is likely to change 153

with the resource levels of the animal. To our knowledge, there are no studies comparing how the 154

mRNA-protein relationship changes across nutrient space. 155

MANUSCRIP

T

ACCEPTED

ACCEPTED MANUSCRIPTHere we address this gap using a model insect, Spodoptera littoralis, (Lepidoptera: Noctuidae), a 156

generalist herbivore. We take a GFN approach, restricting caterpillars to diets that vary in their P:C 157

ratio and energy content, thereby covering a large region of nutrient space. We then challenge the 158

immune system by injecting caterpillars with live or dead bacteria, and measure the expression of 9 159

genes (5 immune, 4 non-immune), and 3 functional immune responses, which are transcribed by two 160

of the immune genes (PPO and lysozyme) in the hemolymph, thus allowing us to associate gene 161

expression and functional immune responses to nutrient intake, and importantly, to assess how well 162

gene expression predicts the immune response specifically associated with those genes under different 163

dietary conditions. 164

165

Material and methods 166

Host and pathogen cultures 167

The Spodoptera littoralis culture was established from eggs collected near Alexandria in Egypt in 168

2011. The colony was reared using single pair matings with around 150 pairs established each 169

generation. Following mating of unrelated adult moths; eggs were laid within 2 days with larvae 170

hatching after a further 3 days. Spodoptera littoralis spend approximately 2 weeks in the larval stage, 171

about 7 days of which are spent in the 5th and 6th instars. Larvae were reared individually from the 172

2nd instar on a semi-artificial wheat germ-based diet (Reeson et al., 1998) in 25 ml polypots until the 173

final larval instar (6th), at which point they were used in the experiments as described below. Insects 174

were maintained at 27°C under a 12:12 light: dark photo regime. 175

Bacteria were originally supplied by the laboratory of Givaudan and colleagues (Montpellier 176

University, France; Xenorhabdus nematophila F1D3 GFP labelled, see (Sicard et al., 2004)). Pure X. 177

nematophila F1D3 stocks were stored at -20°C in Eppendorf tubes (500 µl of X. nematophila F1D3 in 178

nutrient broth with 500 µl of glycerol). Vortexing ensured that all X. nematophila F1D3 cells were 179

coated in glycerol. To revive the stocks for use, 100 µl was added to 10 ml nutrient broth, and 180

incubated at 28°C for up to 48 h (generally stocks had grown sufficiently after 24 hrs). On the day of 181

MANUSCRIP

T

ACCEPTED

ACCEPTED MANUSCRIPTexperimental bacterial challenge, a sub-culture of the stock was carried out, with 1 ml of the original 182

stock added to 10 ml of nutrient broth and placed in a shaker-incubator for approximately 4 h. This 183

ensured that the bacteria were in log phase prior to challenge. Following the sub-culture, a 1 ml 184

sample was checked for purity under the microscope by looking for non-fluorescent cells, which 185

would indicate contamination. The clean sample was then used to produce a serial dilution in nutrient 186

broth from which the total cell count was determined with fluorescence microscopy, using a 187

haemocytometer with improved Neubauer ruling. The remaining culture was diluted with nutrient 188

broth to the appropriate concentration required for the bacterial challenge. The heat-killed treatment 189

group was established by autoclaving the bacteria for 30 min at 121oC. 190

Diet manipulation 191

The aim of the experiments was to tease apart the importance of relative and absolute nutrient effects 192

on immune gene expression and immune protein activity. Therefore, larvae were fed on one of 20 193

chemically-defined diets that varied in both the protein to carbohydrate (P:C) ratio and calorie density 194

based on (Simpson and Abisgold, 1985) (Table 1). This comprised five P:C ratios (5:1, 2:1, 1:1, 1:2, 195

1:5) and four calorie densities (326, 612, 756 and 1112 kJ/100g diet; the remainder of the diet 196

comprising indigestible cellulose (See Table S1 for information about the specific ingredients for each 197

diet). Thus, the 20 diets could be described with respect to the absolute amount of proteins or 198

carbohydrates, by their sum (calorie density), by their ratio (P:C) or by their interaction (P*C). In 199

addition, the absolute amounts of food eaten by the larvae on each diet were recorded so the absolute 200

amount of protein or carbohydrate eaten as opposed to amounts offered could also be used. We were 201

therefore able to define 30 alternative models for describing the relationship between the trait of 202

interest (e.g. Toll expression), and host diet (Table 1). These were then compared using an 203

information theoretic approach by comparing AICc values and other model metrics (Burnham and 204

Anderson, 2003; Whittingham et al., 2006). 205

Bacterial challenge 206

MANUSCRIP

T

ACCEPTED

ACCEPTED MANUSCRIPTXenorhabdus nematophila is a highly pathogenic Gram-negative bacterium. In the wild, this species 207

relies on the entomopathogenic nematode Steinernema carpocapsae, which vectors X. nematophila, to 208

gain access to an insect host, where it rapidly multiplies, generally causing death within 24-48 hours 209

(Georgis et al., 2006; Herbert and Goodrich-Blair, 2007). However, in the lab we can circumvent the 210

requirement for the nematode by injecting X. nematophila directly into the insect haemocoel (Herbert 211

and Goodrich-Blair, 2007). 212

Experiment 1: Within 24 h of moulting to the 6th instar, 400 larvae were divided into 20 groups (n = 213

20 per group), placed individually into 90 mm diameter Petri dishes and provided with ~1.5 g of one 214

of the 20 chemically-defined diets (Table 1). Within each diet, 10 larvae were allocated to the control 215

group and 10 were assigned to the bacteria-challenged group. Following 24 h feeding on the assigned 216

diets (at time, t = 0), 200 larvae were handled then replaced on their diet (control) whilst 200 larvae 217

were injected with 5 µl of a heat killed LD50 dose of X. nematophila (averaging 1272 X. nematophila 218

cells per ml nutrient broth) using a microinjector (Pump 11 Elite Nanomite) fitted with a Hamilton 219

syringe (gauge = 0.5mm). The syringe was sterilised in ethanol prior to use and the challenge was 220

applied to the left anterior proleg. Every 24 h up to 72 h (i.e. 48 h post infection), larvae were 221

transferred individually to clean 90 mm Petri dishes containing 1.5 - 1.8 g of their assigned 222

chemically-defined diet. 96 h after moulting into L6, the larvae had either pupated or were placed on 223

semi-artificial diet until death or pupation. The amount of food eaten each day was determined by 224

weighing the wet mass of the chemically-defined diet provided each day to the caterpillars, as well as 225

weighing uneaten control diets each day (3 control diets per diet). The uneaten diet and control diet 226

were then dried to a constant mass (for approx. 72 h), allowing the consumption per larva to be 227

estimated. 228

Experiment 2: The set up for this experiment was identical to Experiment 1, except that each of the 229

400 larvae was injected with 5 µl of either a heat-killed (control) or live LD50 dose of X. nematophila 230

(averaging 1272 X. nematophila cells per ml nutrient broth). 231

Hemolymph sampling 232

MANUSCRIP

T

ACCEPTED

ACCEPTED MANUSCRIPTFollowing challenge, hemolymph samples were obtained from all caterpillars at 20 h post infection. 233

Hemolymph samples were obtained by piercing the cuticle next to the first proleg near the head with a 234

sterile needle and allowing released hemolymph to bleed directly into an Eppendorf tube. 235

Immediately following hemolymph sampling, 30 µl of fresh hemolymph was added to a sterile ice-236

cooled Eppendorf containing 350 µl of lysis buffer (RLT + Beta mercaptoethanol – 100:1) for later 237

RNA extraction and qPCR analysis (Expts 1 and 2). The remainder of the hemolymph extracted was 238

stored in a separate Eppendorf for further immune assays (Expt 2 only). All hemolymph samples were 239

stored at -80°C prior to processing. 240

Gene expression (Expts 1 and 2) 241

RNA was extracted from hemolymph samples using Qiagen RNeasy mini kit following the 242

manufacturers instructions with a final elution volume of 40 µl. Extracts were quantified using the 243

Nanodrop 2000 and diluted to 0.5 µg/µl for cDNA synthesis. Prior to cDNA synthesis a genomic 244

DNA elimination step was carried out by combining 12 µl RNA (0.5 µg total RNA) plus 2 µl DNA 245

wipeout solution and incubating at 42 °C for 2 min, cDNA synthesis was carried out using Qiagen 246

Quantitect Reverse Transcription kit in a final reaction volume of 20 µl following the manufacturer’s 247

instructions, cDNA synthesis was carried out for 30 min at 42 °C followed by 3 min incubation at 95 248

°C and stored at -20 °C. cDNA was diluted 1:5 for use as a qPCR template. 249

Primers and probes were synthesised by Primer Design and qPCR was performed in a reaction 250

volume of 10 µl with 1x Taqman FAST Universal PCR Master mix (Thermo Fisher), 0.25 µM of each 251

primer, 0.3 µM probe and 2 µl of a 1:5 dilution of cDNA. qPCR was carried on the ABI 7500 FAST, 252

cycling parameters included an initial denaturation at 95 °C for 20 sec followed by 40 cycles of 253

denaturation at 95 °C, 3 sec and annealing at 60 °C for 30 sec. All PCRs were run in duplicate. 254

We selected five immune genes, three from the Toll immune pathway: Toll, Prophenoloxidae (PPO), 255

which is the precursor of the phenoloxidase enzyme (PO), responsible for production of melanin 256

during the encapsulation response, and lysozyme, which produces the antimicrobial lysozyme 257

enzyme, active against Gram positive bacteria. We also selected two genes from the IMD immune 258

MANUSCRIP

T

ACCEPTED

ACCEPTED MANUSCRIPTpathway, Moricin, which produces the AMP Moricin, active against Gram positive and negative 259

bacteria, and Relish, which activates transcription of AMP genes (Ligoxygakis, 2013; Wiesner and 260

Vilcinskas, 2010). We also selected three non-immune genes, Tubulin, a component of the 261

cytoskeleton responsible for organelle and chromosomal movement. Armadillo (b-catenin), which 262

facilitates protein-protein interactions and EF1, an elongation factor facilitates protein synthesis. 263

These genes were selected, due to robust amplification, from a pool of potential endogenous controls 264

that were tested in pilot studies. We also selected Arylphorin, which is primarily characterised as a 265

storage protein (Telfer and Kunkel, 1991), however, it is up-regulated in response to bacterial 266

infection and also in response to non-pathogenic bacteria in the diet of Trichoplusia ni caterpillars 267

(Freitak et al., 2007) and so we did not have an a priori expectation as to its behaviour in this species. 268

269

Lysozyme assays (Expt 2 only) 270

Bacterial agar plates were used to determine lytic activity. These were made by mixing 1.5% water 271

agar and 1.5% freeze-dried Micrococcus luteus (Merck: M3770) potassium phosphate buffer in a 2:1 272

ratio. 10 ml plates of the resulting solution were poured and 2 mm diameter holes punched in each 273

plate. Each hole filled with 1 ml of ethanol saturated with phenylthiourea (PTU), in order to prevent 274

melanisation of the samples. The ethanol evaporates, leaving the PTU in the hole. Following 275

defrosting and vortexing of the stored hemolymph, each well was the filled with 1µl of hemolymph, 276

with two technical replicates per sample. The plates were incubated at 30°C for 24 h, and the clear 277

zones around the samples were measured using digital callipers. Lytic activity (mg/ml) was then 278

calculated from a serial dilution of a hen egg white lysozyme (Merck: 62971; Standard series 0.01, 279

0.05, 0.1, 0.5, 1 and 2 mg per ml in water). 280

Phenoloxidase assays (Expt 2 only) 281

Following defrosting of the hemolymph samples, 10 µl of hemolymph was added to 450 µl of 282

NaCac buffer (1.6g NaCac and 0.556g CaCl2 l-1 sterile distilled water). The solution was then split 283

into two Eppendorfs (each containing 225 µl), in order to carry out assays for both proPO and PO. 284

MANUSCRIP

T

ACCEPTED

ACCEPTED MANUSCRIPTTo one Eppendorf, 25 µl of NaCac buffer was added (PO assay), and to the other, 25 µl of 20 mg 285

per ml chymotrypsin in NaCac buffer was added (proPO activated). The samples were vortexed 286

and incubated at 25 °C for 1 h. 90 µl of each solution was placed in a well of a 96-well microplate 287

with 90 µl of 10 mM dopamine as a substrate. Two technical replicates were carried out per 288

sample. Readings were taken every 15 secs for 10 mins at 490 nm and 25 °C using a Tecan infinite 289

m200pro plate reader with Magellan software (V7.2). This range accounted for the linear stage of 290

the reaction. The maximum rate of reaction was then used as an approximation of PO and proPO 291

level. While many researchers still use L-dopa as a substrate for PO reactions, for insect POs, 292

dopamine is the preferred substrate over L-dopa. It is the natural substrate for insects, it is more 293

soluble than L-dopa and unlike L-dopa, is not subject to spontaneous darkening (Sugumaran, 294

2002). 295

296

Statistical analyses 297

Gene expression 298

All statistical analyses were conducted using the R statistical package version 3.2.2 (R Core Team, 299

2018). Gene expression data were normalised using NORMA-Gene (Heckmann et al., 2011), a data 300

driven approach that normalises gene expression relative to other genes in the dataset rather than to 301

specifically identified reference genes. It is particularly suited to data sets with limited numbers of 302

assayed genes. Normalised gene expression data were then standardized using the mean (µ) and 303

standard deviation (σ) of each trait (Z = (X- µ) ⁄ σ) prior to analysis. The two experiments, run at 304

different times, had only one treatment in common, (1 – handled vs heat-killed bacteria, 2- heat-killed 305

vs live bacteria). For ease of interpretation, we wanted to analyse both experiments in a single model. 306

To test the validity of this approach, we first compared the gene expression, physiological immune 307

response data and the data for the total amount of food consumed across both experiments for the 308

heat-killed treatment only. There was no significant difference between any of the measures across 309

experiments, with the exception of the total amount of food eaten, and expression of the Moricin gene. 310

MANUSCRIP

T

ACCEPTED

ACCEPTED MANUSCRIPTTherefore, all data were analysed in a single model, with the exception of those two response 311

variables, where data from the two experiments were analysed separately. 312

Data were analysed for each gene separately using linear mixed-effects models in the packages lme4 313

(Bates et al., 2015) and lmerTest (Kuznetsova et al., 2017). For each gene, the plate that the samples 314

were run on was included as a random effect. A comparison was made of 90 candidate models for 315

each gene, which comprised 30 models covering different combinations of dietary attributes (Table 316

3), either alone, with bacterial treatments added or with bacterial treatment interacting with the dietary 317

traits. AIC values were corrected for finite sample sizes (AICc) to establish the most parsimonious 318

models including likely nutritional attributes driving the observed data. AICc values and Akaike 319

weights were estimated using the MuMin package (Bartoń, 2018). The relative weight of evidence in 320

favour of one model over another (evidence ratio) is determined by dividing the Akaike weight of one 321

model by another (Burnham and Anderson, 2003). In each case, the residuals from the best model 322

were visually inspected for deviations from normality. Gene expression for Lysozyme, Arylophorin, 323

PPO, EF1 and Tubulin were Tukey transformed prior to reanalysis using the package rcompanion 324

(Mangiafico, 2017). For visualisation of the effects of the immune challenge treatment and diet on 325

gene expression (Figures 2-5), residuals from the null model, containing just the random plate effect 326

(Model 1, Table 3), were plotted as thin plate splines using the package fields (Nychka et al., 2017). 327

Food consumption data were analysed in the same way as the gene expression data, with experiment 328

included as a random effect. 329

330

Physiological immune responses 331

The same approach was taken for the physiological immune measurements, lysozyme, PPO and PO 332

activity, except for these variables, standard linear fixed effects models were run as data were 333

collected in a single experiment. The same set of 90 models as described above were fitted, with the 334

addition of 180 extra models that included the additive and interactive effects of the expression of the 335

relevant gene, after correction for the plate to plate variation (residuals from the null model containing 336

the random effect of plate only) – the lysozyme gene for lysozyme activity and the PPO gene for PPO 337

and PO activity. 338

MANUSCRIP

T

ACCEPTED

ACCEPTED MANUSCRIPTSurvival 339

Time to death was analysed for experiment 2, where larvae were injected with dead or live bacteria 340

only. Data were analysed using Cox’s proportional hazard models in the package (Therneau, 2015). 341

The same sets of models as described above were fitted (Table 3), with the addition of 120 extra 342

models that included the additive and interactive effects of Moricin gene expression, after correction 343

for the plate to plate variation (residuals from the null model containing the random effect of plate 344

only). For visualisation of the effects of the immune challenge treatment on survival (dead vs live 345

bacteria), predicted curves for low and high levels of Moricin gene expression were generated using 346

the package Survminer (Kassambara and Kosinski, 2018) using ggplot2 (Wickham, 2016.). The 347

effects of diet on time to death were plotted as thin plate splines using the package fields (Nychka et 348

al., 2017). 349

350

Results 351

How does consumption vary across diets and bacterial challenge treatments? 352

The total amount of food consumed varied across the diets. For experiment 1, comparing handled 353

caterpillars versus those injected with heat-killed bacteria, the best model predicting consumption was 354

model 30 (Pe*Ce+Pe2+Ce2), but this was indistinguishable from the same model that included the 355

additive effects of treatment (Treatment+ Pe*Ce+Pe2+Ce2, delta=1.34). 356

For experiment 2, comparing dead and live bacterial injections, the best model predicting 357

consumption was model 20 (Co*R+Co2+R2), but as for the handled versus dead treatments in 358

experiment 1, this model was indistinguishable from the same model that included the additive effects 359

of treatment (Treatment+Co*R+Co2+R2, delta=0.51). 360

For all treatment groups, it can be seen that consumption tended to increase as the calorie density of 361

the diet decreased (Figure 1a,b,d,e), suggesting that food dilution constrained caterpillars from being 362

able to take in sufficient nutrients, as expected, and that on the more calorie-dense diets caterpillars 363

over-consumed nutrients. However, this increase in total consumption was more extreme on the high-364

MANUSCRIP

T

ACCEPTED

ACCEPTED MANUSCRIPTprotein than on the low-protein diets, suggesting that caterpillars were willing to overeat protein to 365

gain limiting carbohydrates. 366

Overall consumption tended to decrease with treatment - dead-bacteria treated caterpillars ate less 367

than handled, and live-bacteria treated caterpillars ate less than dead-bacteria treated (Figure 1a vs b 368

and d vs e). However, inspection of the intake arrays (Figures 1c,e), suggests that consumption was 369

most reduced in both dead and live bacteria treatments on the highest protein diets. 370

371

How does immune gene expression vary across diets and bacterial challenge treatments? 372

For the immune genes (Toll, PPO, Lysozyme, Moricin and Relish), injection with dead bacteria 373

resulted in up-regulation of gene expression relative to handled caterpillars (Figure 2). In contrast, 374

injection with live bacteria either did not up-regulate gene expression relative to controls (Toll, PPO 375

and Lysozyme), or did not up-regulate it as strongly (Moricin and Relish) (Figure 2). For the non-376

immune genes (Arylophorin, EF1, Armadillo and Tubulin), the variation in expression levels was 377

lower; for Arylophorin, EF1 and Armadillo, live bacteria triggered the down-regulation of gene 378

expression relative to handled caterpillars, whilst there was no effect for Tubulin (Figure 2). For 379

Arylophorin, Armadillo and Tubulin, injection with dead bacteria up-regulated gene expression 380

relative to handled caterpillars but there was no effect for EF1 (Figure 2). The best supported model 381

for every gene tested was model 30, with the bacterial treatment interacting with the amount of 382

protein and carbohydrate eaten (Treatment*(Pe*Ce+Pe2+Ce2)). However, although the fit of these 383

models was generally good (r2 > 0.26-0.86), with the exception of Moricin, the amount of variation 384

explained by the fixed part of the model was very low (r2 < 0.12; Table 4; Figures 3-5). This means 385

that the majority of the variation in gene expression was caused by variation across plates. For 386

Moricin, when comparing the handled and dead treatments, 74% of the variation explained by the 387

model was explained by the fixed terms due to the massive up-regulation of Moricin in the dead-388

bacteria injected larvae (Figures 2, 3a,b). The difference between the dead and live treatment groups 389

was much smaller and comparable to the other immune genes (Table 4, Figures 3c,d) 390

MANUSCRIP

T

ACCEPTED

ACCEPTED MANUSCRIPTVariation in the expression of all of the genes was explained by main and interactive effects of the 391

amount of protein and carbohydrate eaten, and in interaction with the bacterial treatment, suggesting 392

that the response to diet for each gene differed across treatments. A visualisation of these response 393

surfaces (Figures 3-5) shows that, for the immune genes, whilst there is general up-regulation between 394

handled and dead bacterial challenges, the response surfaces are fairly flat, i.e. diet does not explain 395

much variation in gene expression. However, for the live challenge, expression tends to peak at 396

moderate protein but high carbohydrate intake, which corresponds to the highest intakes on the 33% 397

protein diet for Toll, PPO, Lysozyme and Relish, and on the 17% protein diet for Moricin (Figures 398

3,4). In contrast, the non-immune genes (Arylphorin, EF1, Armadillo and Tubulin), show a 399

consistently weak response to the dietary manipulation, with much flatter surfaces on average than 400

those shown by the immune genes (compare Figure 4 with Figure 5). 401

402

Does immune gene expression predict physiological immune responses? 403

For the Lysozyme and PPO genes, we simultaneously measured functional lytic and PPO (and PO) 404

activity in the hemolymph, allowing us to determine how well gene expression predicts the functional 405

immune response. We had lytic and PO data only for Experiment 2, where larvae were challenged 406

with live or dead bacteria. 407

For PPO activity, AICc could not discriminate between several of the diet models, with seven being 408

equally well supported (delta < 2; Table 5). Of these models, the top six contained protein and protein 409

squared with additive or interactive effects of bacterial treatment or gene expression (Table 5). For the 410

models that included treatment, the estimates show that PPO activity was increased with live bacterial 411

infection. For PO activity, AICc could not discriminate between 11 different models (delta < 2; Table 412

6). However, the top three models were the same as for PPO, with protein plus protein squared with 413

additive or interactive effects of PPO gene expression. Only two of the models contained treatment 414

effects and both in interaction with diet components. For lytic activity in the hemolymph, three 415

models were equally well supported, all of which contained Lysozyme gene expression interacting 416

MANUSCRIP

T

ACCEPTED

ACCEPTED MANUSCRIPTwith dietary components, which were either protein and protein squared, as for PO and PPO, or the 417

P:C ratio (Table 7); none of the models contained treatment, suggesting that lysozyme activity is up-418

regulated in response to the presence of bacteria and not whether they are alive or dead. As for gene 419

expression, the overall explanatory power of the models was quite low, (r2< 0.12; Tables 5-7). 420

For ease of comparison, all 3 physiological immune traits were plotted against the protein content of 421

the diet, as this model was common to all three traits, and the expression of the relevant gene, which 422

featured in the majority of the selected models (Tables 5-7). The effect of treatment was excluded as it 423

did not feature in the majority of the models. For each trait, activity in the hemolymph tended to 424

increase with gene expression, as we might expect, but this was strongly moderated by the protein 425

content of the diet (Figure 6). For PO and PPO activity, on low protein diets enzyme activity was low 426

and there was little correspondence between gene expression and the physiological response, but as 427

the protein content of the diet increased, this relationship became more linear (Figure 6a,b). For lytic 428

activity the pattern was different in that enzyme activity increased strongly with the protein and less 429

strongly with lysozyme gene expression up to about 45% protein, thereafter there was consistently 430

high lytic activity across all levels of gene expression (Figure 6c). 431

432

Does immune gene expression predict survival? 433

Survival was reduced in the live bacterial treatment group relative to those injected with dead bacteria 434

(Hazard ratios 1.25-1.31 for models without treatment interactions, Table 3), however, this effect was 435

moderated by Moricin expression (Figure 7 a,b). In the dead-bacteria treatment group, Moricin did not 436

explain time to death, but in the live-bacteria treatment group, larvae with high levels of Moricin 437

expression had an increased risk of death relative to those with low expression (Figure 7 a,b; Hazard 438

ratios 1.20-1.24 for models without GE interactions, Table 3). Of the top 5 models, 4 included the 439

additive and interactive effects of protein and carbohydrate eaten as well as their squared terms (Table 440

3). To visualise the effects of diet on survival we plotted thin-plate splines for time to death against 441

the amount of protein and carbohydrate consumed. The patterns differ between dead and live bacterial 442

MANUSCRIP

T

ACCEPTED

ACCEPTED MANUSCRIPTtreatments. Time to death is overall shorter in the live treatment (note the shift of colour towards 443

orange and blue). However, whilst time to death is affected by both protein and carbohydrate 444

consumption in the dead treatment, with peak survival on high protein/low carbohydrate and vice 445

versa, in the live treatment, time to death appears to be solely explained by protein availability (note 446

the near-vertical contours). Low-protein diets resulted in the most rapid deaths and high-protein diets 447

extended the time to death. 448

449

Discussion 450

Previous work has shown that immune responses can be strongly affected by the amount and/or 451

balance of nutrients in the diet e.g. (Fernandes et al., 1976; Ingram et al., 1995; Kristan, 2008; Le 452

Couteur et al., 2015; Lee et al., 2006; Nayak et al., 2009; Povey et al., 2009; Ritz and Gardner, 2006; 453

Wallace et al., 1999; White et al., 2017). However, most of these studies covered only a relatively 454

small region of nutrient space (Fernandes et al., 1976; Ingram et al., 1995; Lee et al., 2006; Nayak et 455

al., 2009; Povey et al., 2009; Ritz and Gardner, 2006; White et al., 2017) and/or only tested innate 456

responses (e.g.(Fernandes et al., 1976; Ingram et al., 1995; Le Couteur et al., 2015; Lee et al., 2006; 457

Nayak et al., 2009; Povey et al., 2009; White et al., 2017) or the response to an artificial pathogen or 458

immune stimulant (Cotter et al., 2011b). Here we addressed this gap by looking at both gene 459

expression, functional immune responses and survival after both dead and live pathogen challenges 460

over a broad region of nutrient space. Our major findings are that whilst functional immune responses 461

(PPO, PO and lytic activity in the hemolymph) change as expected in response to the dietary 462

manipulation, showing a clear elevation as the protein content of the diet increases, gene expression is 463

much less predictable (Figures 3,4). Despite this, expression of the PPO and Lysozyme genes did 464

predict PPO/PO and Lysozyme activity in the hemolymph, but this relationship was strongly 465

dependent on the amount of protein in the diet (Figure 6), suggesting that using immune gene 466

expression as an indicator of the efficacy of the immune response may be reliable only under specific 467

dietary conditions. Furthermore, expression of the most responsive gene to infection (Moricin) 468

strongly modulated survival, with high levels of expression resulting in reduced survival after 469

MANUSCRIP

T

ACCEPTED

ACCEPTED MANUSCRIPTbacterial infection, suggesting that expression is a marker of bacterial load or ‘sickness’ as opposed to 470

an indication of a robust immune response. 471

Our dietary manipulation was successful in inducing caterpillars to consume over a large region of 472

nutrient space, allowing us to independently assess the effects of macronutrient composition and the 473

calorie content of the diet on immunity. There was evidence for compensatory feeding; caterpillars 474

did not consume the same amount of every diet. As expected, caterpillars ate more as the calorie 475

density of the food decreased (Figure 1), but this varied across diets, such that consumption was 476

highest on the high protein diets, suggesting that caterpillars were willing to over-eat protein to gain 477

limiting carbohydrates. However, as has been found in previous studies (Adamo, 1998; Adamo et al., 478

2007; Exton, 1997; Lennie, 1999; Povey et al., 2014), we found some evidence for illness-induced 479

anorexia. Caterpillars injected with live X. nematophila showed suppressed food consumption across 480

all diets (Figure 1e – note the shift of colours towards oranges and blues). Interestingly, injection with 481

dead X. nematophila did not induce this response, which suggests that it is not the triggering of an 482

immune response that causes this change in consumption, but the presence of an actively replicating 483

pathogen. This reduction in consumption was also consistent across diets, with infected caterpillars, 484

on average, consuming just 77% of the food consumed by healthy caterpillars (Figure 1c). 485

In insects, two major pathways are triggered in response to microbial infection; typically, genes in the 486

Toll pathway respond to infection by fungi and Gram-positive bacteria, whilst genes in the IMD 487

pathway respond to Gram-negative bacteria (Broderick et al., 2009). Moricin and Lysozyme are 488

triggered by Toll in Lepidoptera (e.g. (Zhong et al., 2016), but Moricin has also been shown to 489

respond to Gram-negative bacteria and so may also be triggered by IMD (Hara and Yamakawa, 490

1995). Of the 5 immune genes we tested, only the IMD genes, Moricin and Relish, were significantly 491

up-regulated in response to infection with both dead and live bacteria. For the Toll genes (Toll, PPO 492

and Lysozyme), gene expression was up-regulated by dead bacteria but not by live bacteria (Figure 2). 493

However, even for Moricin and Relish, up-regulation was much stronger in response to dead than live 494

bacteria. This may reflect a general down-regulation of gene expression during an active infection, as 495

the non-immune genes typically show reduced gene expression in response to the live infection 496

MANUSCRIP

T

ACCEPTED

ACCEPTED MANUSCRIPTcompared to the controls. This may be driven by the illness-induced anorexia, with reduced 497

consumption resulting in lower metabolic activity and consequently lower gene expression. However, 498

there is evidence that X. nematophila can inhibit Cecropin, Attacin and Lysozyme gene expression (Ji 499

and Kim, 2004; Park et al., 2007). It may be that, rather than specifically inhibiting AMP gene 500

expression, X. nematophila inhibits the expression of all genes. 501

As Moricin was most strongly up-regulated in response to infection, we tested how its expression 502

correlated with time to death in challenged caterpillars (dead vs live injection, Expt 2 only). Whilst 503

Moricin expression had negligible effects on survival in the dead bacterial treatment, high levels of 504

expression were indicative of an increased risk of death after live infection. Thus, high expression 505

levels were not a good indicator of immune capacity, but rather signalled heavy bacterial loads or low 506

tolerance. Distinguishing between these hypotheses would require data on bacterial loads at different 507

time points after infection. Survival was also strongly affected by diet, with the longest survival times 508

occurring on the highest protein diets after live-bacteria infection. High-protein diets have been 509

implicated in increased survival after viral infection in this species (Lee et al., 2006) and after either 510

bacterial or viral infection in the congener, Spodoptera exempta (Povey et al., 2009; Povey et al., 511

2014). However, none of these diets are associated with the highest gene expression for any immune 512

gene, suggesting that high-protein diets may reduce the burden of infection via mechanisms other than 513

improving the immune response. 514

X. nematophila is a Gram-negative bacterium, and is clearly triggering Moricin and Relish expression, 515

but as Toll is only marginally up-regulated in response, it is probably the IMD pathway that is 516

controlling this response. Another possible explanation for why live bacteria appear to trigger a down-517

regulation of gene expression is that our sampling protocol (20 h post-challenge) did not allow us to 518

catch peak expression levels (note that bacterial loads tend to peak in S. littoralis at around 24h). 519

Expression of lysozyme and PPO in the Glanville fritillary butterfly was not up-regulated 24 h after 520

injection with M. luteus cells (Woestmann et al., 2017) , whilst in the silkworm, up-regulation of 521

lysozyme in response to fungal infection occurred in two peaks, from 9-18 h, and then between 30 and 522

48 h (Hou et al., 2014) . This may be a fungal-specific response, or it might mean that we would have 523

MANUSCRIP

T

ACCEPTED

ACCEPTED MANUSCRIPTseen higher gene expression had we assayed over an extended time period. It is also possible that the 524

timing of gene expression peaks earlier after live, rather than dead bacterial injection, further studies 525

would be required to elucidate the time course of gene expression for the different genes to be certain 526

of this. However, as non-immune genes also appear to follow the same pattern, reduced expression in 527

response to live vs dead bacteria, the hypothesis that infection results in down-regulation of gene 528

expression in general is a reasonable assumption. 529

Arylphorin is primarily characterised as a storage protein (Telfer and Kunkel, 1991), however, it is 530

up-regulated in response to bacterial infection and also in response to non-pathogenic bacteria in the 531

diet of Trichoplusia ni caterpillars (Freitak et al., 2007). It has been shown to bind to fungal conidia in 532

Galleria mellonella hemolymph, potentially working in coordination with antimicrobial peptides 533

(Fallon et al., 2011). The lack of up-regulation here may be due to the use of a Gram-negative 534

bacterial challenge; the up-regulation in T. ni was in response to a mixture of E. coli (G-ve) and 535

Micrococcus luteus (G+ve), so it is not clear if both or just one of the bacteria caused the response. 536

Another possibility is that Arylphorin levels are already expressed at maximal levels and cannot be 537

further up-regulated. In T. ni caterpillars, Arylphorin is the most abundant protein in the hemolymph 538

during the final instar (Kunkel et al. 1990). Its levels are known to increase throughout the final instar 539

in Spodoptera litura (Yoshiga et al., 1997), and the point at which gene expression was measured here 540

was 48-72 h into the final instar, which is shortly before pupation. The pattern of gene regulation for 541

Arylphorin looks more like that shown by the non-immune genes, with little or no up-regulation in 542

response to dead bacteria and down-regulation in response to live bacteria. Further studies would be 543

required to assess the role of Arylphorin as a putative immune gene in this species. 544

For two of the immune genes, PPO and Lysozyme, we were able to simultaneously measure the 545

activity of the relevant protein in the hemolymph as a measure of the functional immune response. 546

Thus, we were able to assess how well gene expression predicts functional immune activity and 547

whether this relationship changes with the diet. Here, we found that for each functional immune 548

response, PPO activity, PO activity and lysozyme activity, expression of the relevant gene does 549

predict the response, but only at certain intakes of protein (Figure 6). For example, PPO and PO 550

MANUSCRIP

T

ACCEPTED

ACCEPTED MANUSCRIPTactivity increase linearly with the expression of the PPO gene, but only above ~30% dietary protein 551

(Figure 6). This suggests that the availability of dietary protein limits the translation of PPO mRNA 552

into PPO protein, and the activation of PPO into PO. In contrast, the expression of the gene is not 553

limited by protein availability, and so gene expression can be high when dietary protein is low, but it 554

is ineffective as it does not result in a comparable functional immune response. The lytic response is 555

also affected by dietary protein, however, in this case, the relationship between gene expression and 556

lytic activity is consistently weak and above 45% protein maximal lytic activity is achieved at low 557

gene expression, and increased expression does not improve the response. As for PPO, this suggests 558

that protein limits the translation of lysozyme up to about 45% protein. 559

These results are not surprising when you consider the costs associated with the production of protein. 560

It is estimated that only 10% of the energetic costs of protein production are spent on transcription; 561

translation is much more energetically expensive and relies on the availability of amino acids to build 562

the relevant protein (Warner, 1999). It is likely, therefore, that whilst transcription of immune genes 563

might still be up-regulated in response to infection under low protein conditions, the translation of the 564

protein might be reduced, impairing the correlation between mRNA and protein abundance. It is also 565

possible that gene expression would be a better predictor of the functional response at different time 566

points, if there is a lag between gene expression and protein translation. Again, this would require 567

further investigation. However, given the much stronger relationship between the physiological 568

immune responses and protein availability, it still seems likely that the relationship between the two 569

will differ across diets. Our results suggest that caution should be used when interpreting gene 570

expression as a measure of “investment” into a particular trait, or as a measure of the strength of a 571

particular immune response. It is surprisingly common in ecological studies for gene expression to be 572

used in this way without any attempt to correlate the expression of a gene with the production of the 573

functional protein (Zylberberg, 2019). If dietary protein levels are limiting, then gene expression may 574

be a poor indicator of the immune capacity of an animal. Here we have tested this just with immune 575

genes for which we have good functional assays of the active protein, but it seems likely that this 576

MANUSCRIP

T

ACCEPTED

ACCEPTED MANUSCRIPTwould also be true of other classes of genes, for which gene expression is routinely used as an 577

indicator of an organism’s investment. 578

In summary, as expected, immune challenge with a live Gram-negative bacterium up-regulated 579

immune genes in the IMD pathway, though all immune genes were up-regulated to a certain extent by 580

the challenge with dead bacteria. While functional immune responses (PO, PPO and Lysozyme) 581

typically improved with the protein content of the diet, gene expression varied non-linearly with diet 582

composition. However, the expression of PPO and Lysozyme genes predicted PPO/PO and Lysozyme 583

activity, but only when the availability of dietary protein was not limiting, suggesting that using gene 584

expression as an indicator of investment in a trait is unlikely to be reliable, unless its relationship with 585

diet is known. 586

Acknowledgements 587

This study was funded by a standard research grant awarded to KW, JAS and SJS by the 588

UK’s Biotechnology and Biological Sciences Research Council (BB/I02249X/1). SCC was supported 589

by a Natural Environment Research Council Fellowship (NE/H014225/2). We are grateful to Alain 590

Givaudan and colleagues (Montpellier University, France) for gifting the Xenorhabdus 591

nematophila F1D3 bacteria, and to Esmat Hegazi for his support with Spodoptera littoralis. We thank 592

Phill Nott for technical assistance. We thank the editor and two anonymous referees whose comments 593

greatly improved the manuscript. 594

595

Author contributions 596

KW, JAS, SCC, FP and SJS conceived the idea, RH, CER, JR, JAS & YT carried out the 597

experiments, SCC analysed the data and wrote the first draft of the paper, all authors commented on 598

and approved the final manuscript. 599

600

601

MANUSCRIP

T

ACCEPTED

ACCEPTED MANUSCRIPT 602

References 603

Adamo, S.A., 1998. Feeding suppression in the tobacco hornworm, Manduca sexta: costs and 604

benefits to the parasitic wasp Cotesia congregata. Can. J. Zool.-Rev. Can. Zool. 76, 1634-605

1640. 606

Adamo, S.A., Fidler, T.L., Forestell, C.A., 2007. Illness-induced anorexia and its possible 607

function in the caterpillar, Manduca sexta. Brain Behav. Immun. 21, 292-300. 608

Alonso-Alvarez, C., Tella, J.L., 2001. Effects of experimental food restriction and body-mass 609

changes on the avian T-cell-mediated immune response. Can. J. Zool.-Rev. Can. Zool. 79, 610

101-105. 611

Bartoń, K., 2018. MuMIn: Multi-Model Inference, https://CRAN.R-612

project.org/package=MuMIn. 613

Bates, D., Maechler, M., Bolker, B., Walker, S., 2015. Fitting Linear Mixed-Effects Models 614

Using lme4. Journal of Statistical Software 67, 1-48. 615

Broderick, N.A., Welchman, D.P., Lemaitre, B., 2009. Recognition and response to microbial 616

infection in Drosophila. Oxford University Press: New York, NY, USA. 617

Brown, M.J.F., Loosli, R., Schmid-Hempel, P., 2000. Condition-dependent expression of 618

virulence in a trypanosome infecting bumblebees. Oikos 91, 421-427. 619

Burnham, K.P., Anderson, D.R., 2003. Model selection and multimodel inference: a practical 620

information-theoretic approach. Springer Science & Business Media. 621

Cotter, S., Simpson, S., Raubenheimer, D., Wilson, K., 2011a. Macronutrient balance 622

mediates trade-offs between immune function and life history traits. Functional Ecology 25, 623

186-198. 624

MANUSCRIP

T

ACCEPTED

ACCEPTED MANUSCRIPT

Cotter, S.C., Simpson, S.J., Raubenheimer, D., Wilson, K., 2011b. Macronutrient balance 625

mediates trade-offs between immune function and life history traits. Functional Ecology 25, 626

186-198. 627

Cunningham-Rundles, S., McNeeley, D.F., Moon, A., 2005. Mechanisms of nutrient 628

modulation of the immune response. J. Allergy Clin. Immunol. 115, 1119-1128. 629

Edfors, F., Danielsson, F., Hallström, B.M., Käll, L., Lundberg, E., Pontén, F., Forsström, B., 630

Uhlén, M., 2016. Gene‐specific correlation of RNA and protein levels in human cells and 631

tissues. Molecular Systems Biology 12. 632

Exton, M.S., 1997. Infection-induced anorexia: Active host defence strategy. Appetite 29, 633

369-383. 634

Fallon, J.P., Troy, N., Kavanagh, K., 2011. Pre-exposure of Galleria mellonella larvae to 635

different doses of Aspergillus fumigatus conidia causes differential activation of cellular and 636

humoral immune responses. Virulence 2, 413-421. 637

Fanson, B.G., Weldon, C.W., Perez-Staples, D., Simpson, S.J., Taylor, P.W., 2009. Nutrients, 638

not caloric restriction, extend lifespan in Queensland fruit flies (Bactrocera tryoni). Aging 639

Cell 8, 514-523. 640

Fernandes, G., Yunis, E.J., Good, R.A., 1976. Influence of protein restriction on immune 641

functions in nzb mice. J. Immunol. 116, 782-790. 642

Field, C.J., Johnson, I.R., Schley, P.D., 2002. Nutrients and their role in host resistance to 643

infection. Journal of Leukocyte Biology 71, 16-32. 644

Freitak, D., Wheat, C.W., Heckel, D.G., Vogel, H., 2007. Immune system responses and 645

fitness costs associated with consumption of bacteria in larvae of Trichoplusia ni. BMC 646

Biology 5. 647

Gedeon, T., Bokes, P., 2012. Delayed protein synthesis reduces the correlation between 648

mRNA and protein fluctuations. Biophys. J. 103, 377-385. 649

MANUSCRIP

T

ACCEPTED

ACCEPTED MANUSCRIPT

Georgis, R., Koppenhöfer, A., Lacey, L., Bélair, G., Duncan, L., Grewal, P., Samish, M., 650

Tan, L., Torr, P., Van Tol, R., 2006. Successes and failures in the use of parasitic nematodes 651

for pest control. Biological Control 38, 103-123. 652

Gonzalez-Santoyo, I., Cordoba-Aguilar, A., 2012. Phenoloxidase: a key component of the 653

insect immune system. Entomologia Experimentalis Et Applicata 142, 1-16. 654

Graham, R.I., Deacutis, J.M., Pulpitel, T., Ponton, F., Simpson, S.J., Wilson, K., 2014. 655

Locusts increase carbohydrate consumption to protect against a fungal biopesticide. Journal 656

of Insect Physiology 69, 27-34. 657

Gupta, A.P., 1991. Immunology of Insects and Other Arthropods. CRC Press, Boca Raton, 658

Florida 659

Hara, S., Yamakawa, M., 1995. Moricin, a novel type of antibacterial peptide isolated from 660

the silkworm, Bombyx mori. J. Biol. Chem. 270, 29923-29927. 661

Heckmann, L.-H., Sørensen, P.B., Krogh, P.H., Sørensen, J.G., 2011. NORMA-Gene: A 662

simple and robust method for qPCR normalization based on target gene data. BMC 663

Bioinformatics 12, 250. 664

Herbert, E.E., Goodrich-Blair, H., 2007. Friend and foe: the two faces of Xenorhabdus 665

nematophila. Nat. Rev. Microbiol. 5, 634. 666

Hinnebusch, A.G., Natarajan, K., 2002. Gcn4p, a master regulator of gene expression, is 667

controlled at multiple levels by diverse signals of starvation and stress. Eukaryotic cell 1, 22-668

32. 669

Hou, C., Qin, G., Liu, T., Geng, T., Gao, K., Pan, Z., Qian, H., Guo, X., 2014. Transcriptome 670

analysis of silkworm, Bombyx mori, during early response to Beauveria bassiana challenges. 671

PLoS One 9, e91189. 672

Ingram, K.G., Croy, B.A., Woodward, B.D., 1995. Splenic natural killer cell activity in 673

wasted, protein-energy malnourished weanling mice. Nutr. Res. 15, 231-243. 674

MANUSCRIP

T

ACCEPTED

ACCEPTED MANUSCRIPT

Jackson, J.A., Begon, M., Birtles, R., Paterson, S., Friberg, I.M., Hall, A., Ralli, C., Turner, 675

A., Zawadzka, M., Bradley, J.E., 2011. The analysis of immunological profiles in wild 676

animals: a case study on immunodynamics in the field vole, Microtus agrestis. Molecular 677

Ecology 20, 893-909. 678

Jensen, K., McClure, C., Priest, N.K., Hunt, J., 2015. Sex-specific effects of protein and 679

carbohydrate intake on reproduction but not lifespan in Drosophila melanogaster. Aging Cell 680

14, 605-615. 681

Ji, D., Kim, Y., 2004. An entomopathogenic bacterium, Xenorhabdus nematophila, inhibits 682

the expression of an antibacterial peptide, cecropin, of the beet armyworm, Spodoptera 683

exigua. Journal of insect physiology 50, 489-496. 684

Jovanovic, M., Rooney, M.S., Mertins, P., Przybylski, D., Chevrier, N., Satija, R., Rodriguez, 685

E.H., Fields, A.P., Schwartz, S., Raychowdhury, R., Mumbach, M.R., Eisenhaure, T., Rabani, 686

M., Gennert, D., Lu, D., Delorey, T., Weissman, J.S., Carr, S.A., Hacohen, N., Regev, A., 687

2015. Dynamic profiling of the protein life cycle in response to pathogens. Science 347. 688

Kassambara, A., Kosinski, M., 2018. survminer: Drawing Survival Curves using 'ggplot2', 689

https://CRAN.R-project.org/package=survminer. 690

Kristan, D.M., 2008. Calorie restriction and susceptibility to intact pathogens. Age 30, 147-691

156. 692

Kuvibidila, S., Yu, L., Ode, D., Warrier, P., 1993. The immune response in protein-energy 693

malnutrition and single nutrient deficiencies., in: Klurfeld, D. (Ed.), Nutrition and 694

Immunology. Plenum Press., New York and London, pp. 121-155. 695

Kuznetsova, A., Brockhoff, P., Christensen, R., 2017. lmerTest Package: Tests in Linear 696

Mixed Effects Models. Journal of Statistical Software 82, 1-26. 697

Le Couteur, D.G., Tay, S.S., Solon-Biet, S., Bertolino, P., McMahon, A.C., Cogger, V.C., 698

Colakoglu, F., Warren, A., Holmes, A.J., Pichaud, N., Horan, M., Correa, C., Melvin, R.G., 699

MANUSCRIP

T

ACCEPTED

ACCEPTED MANUSCRIPT

Turner, N., Ballard, J.W.O., Ruohonen, K., Raubenheimer, D., Simpson, S.J., 2015. The 700

Influence of Macronutrients on Splanchnic and Hepatic Lymphocytes in Aging Mice. J. 701

Gerontol. Ser. A-Biol. Sci. Med. Sci. 70, 1499-1507. 702

Lee, K.P., Cory, J.S., Wilson, K., Raubenheimer, D., Simpson, S.J., 2006. Flexible diet 703

choice offsets protein costs of pathogen resistance in a caterpillar. Proceedings of the Royal 704

Society B-Biological Sciences 273, 823-829. 705

Lee, K.P., Simpson, S.J., Clissold, F.J., Brooks, R., Ballard, J.W.O., Taylor, P.W., Soran, N., 706

Raubenheimer, D., 2008. Lifespan and reproduction in Drosophila: New insights from 707

nutritional geometry. Proceedings of the National Academy of Sciences of the United States 708

of America 105, 2498-2503. 709

Lennie, T.A., 1999. Anorexia in response to acute illness. Heart Lung 28, 386-401. 710

Ligoxygakis, P., 2013. Genetics of Immune Recognition and Response in Drosophila host 711

defense, in: Friedmann, T., Dunlap, J.C., Goodwin, S.F. (Eds.), Advances in Genetics, Vol 712

83, pp. 71-97. 713

Liu, Y., Beyer, A., Aebersold, R., 2016. On the dependency of cellular protein levels on 714

mRNA abundance. Cell 165, 535-550. 715

Maklakov, A.A., Simpson, S.J., Zajitschek, F., Hall, M.D., Dessmann, J., Clissold, F., 716

Raubenheimer, D., Bonduriansky, R., Brooks, R.C., 2008. Sex-specific fitness effects of 717

nutrient intake on reproduction and lifespan. Curr. Biol. 18, 1062-1066. 718

Mangiafico, S., 2017. rcompanion: Functions to Support Extension Education Program 719

Evaluation, https://CRAN.R-project.org/package=rcompanion. 720

Mokdad, A.H., Bowman, B.A., Ford, E.S., Vinicor, F., Marks, J.S., Koplan, J.P., 2001. The 721

continuing epidemics of obesity and diabetes in the United States. Jama-Journal of the 722

American Medical Association 286, 1195-1200. 723

MANUSCRIP

T

ACCEPTED

ACCEPTED MANUSCRIPT

Moret, Y., Schmid-Hempel, P., 2000. Survival for immunity: The price of immune system 724

activation for bumblebee workers. Science 290, 1166-1168. 725

Muller, O., Krawinkel, M., 2005. Malnutrition and health in developing countries. Canadian 726

Medical Association Journal 173, 279-286. 727

Nayak, B.N., Friel, J.K., Rempel, C.B., Jones, P.J., 2009. Energy-restricted diets result in 728

higher numbers of CD4+, CD8+, immunoglobulins (A, M, and G), and CD45RA cells in 729

spleen and CD4+, immunoglobulin A, and CD45RA cells in colonic lamina propria of rats. 730

Nutr. Res. 29, 487-493. 731

Nychka, D., Furrer, R., Paige, J., Sain, S., 2017. fields: Tools for spatial data., 732

http://doi.org/10.5065/D6W957CT. 733

Park, Y., Herbert, E.E., Cowles, C.E., Cowles, K.N., Menard, M.L., Orchard, S.S., Goodrich-734

Blair, H., 2007. Clonal variation in Xenorhabdus nematophila virulence and suppression of 735

Manduca sexta immunity. Cell Microbiol. 9, 645-656. 736

Paskewitz, S., Riehle, M.A., 1994. Response of Plasmodium Refractory and Susceptible 737

Strains of Anopheles gambiae to Inoculated Sephadex Beads. Developmental and 738

Comparative Immunology 18, 369-375. 739

Ponton, F., Lalubin, F., Fromont, C., Wilson, K., Behm, C., Simpson, S.J., 2011. Hosts use 740

altered macronutrient intake to circumvent parasite-induced reduction in fecundity. Int. J. 741

Parasit. 41, 43-50. 742

Ponton, F., Wilson, K., Holmes, A.J., Cotter, S.C., Raubenheimer, D., Simpson, S.J., 2013. 743

Integrating nutrition and immunology: A new frontier. Journal of Insect Physiology 59, 130-744

137. 745

Povey, S., Cotter, S.C., Simpson, S.J., Lee, K.P., Wilson, K., 2009. Can the protein costs of 746

bacterial resistance be offset by altered feeding behaviour? Journal of Animal Ecology 78, 747

437-446. 748

MANUSCRIP

T

ACCEPTED

ACCEPTED MANUSCRIPT

Povey, S., Cotter, S.C., Simpson, S.J., Wilson, K., 2014. Dynamics of macronutrient self-749

medication and illness-induced anorexia in virally infected insects. Journal of Animal 750

Ecology 83, 245-255. 751

Reeson, A.F., Wilson, K., Gunn, A., Hails, R.S., Goulson, D., 1998. Baculovirus resistance in 752

the noctuid Spodoptera exempta is phenotypically plastic and responds to population density. 753

Proc. R. Soc. Lond. Ser. B-Biol. Sci. 265, 1787-1791. 754

Ritz, B.W., Gardner, E.M., 2006. Malnutrition and energy restriction differentially affect 755

viral immunity. J. Nutr. 136, 1141-1144. 756

Samartin, S., Chandra, R.K., 2001. Obesity, overnutrition and the immune system. Nutr. Res. 757

21, 243-262. 758

Sicard, M., Brugirard-Ricaud, K., Pagès, S., Lanois, A., Boemare, N.E., Brehélin, M., 759

Givaudan, A., 2004. Stages of infection during the tripartite interaction between Xenorhabdus 760

nematophila, its nematode vector, and insect hosts. Applied and environmental microbiology 761

70, 6473-6480. 762

Simpson, S.J., Abisgold, J.D., 1985. Compensation by locusts for changes in dietary nutrients 763

- behavioral mechanisms. Physiological Entomology 10, 443-452. 764

Simpson, S.J., Raubenheimer, D., 1995. The geometric analysis of feeding and nutrition - a 765

users guide. Journal of Insect Physiology 41, 545-553. 766

Simpson, S.J., Sibly, R.M., Lee, K.P., Behmer, S.T., Raubenheimer, D., 2004. Optimal 767

foraging when regulating intake of multiple nutrients. Animal Behaviour 68, 1299-1311. 768

Solon-Biet, S.M., Walters, K.A., Simanainen, U.K., McMahon, A.C., Ruohonen, K., Ballard, 769

J.W.O., Raubenheimer, D., Handelsman, D.J., Le Couteur, D.G., Simpson, S.J., 2015. 770

Macronutrient balance, reproductive function, and lifespan in aging mice. Proceedings of the 771

National Academy of Sciences of the United States of America 112, 3481-3486. 772

MANUSCRIP

T

ACCEPTED

ACCEPTED MANUSCRIPT

Sugumaran, H., 2002. Comparative biochemistry of eumelanogenesis and the protective roles 773

of phenoloxidase and melanin in insects. Pigm. Cell. Res. 15, 2-9. 774

Telfer, W.H., Kunkel, J.G., 1991. The Function and Evolution of Insect Storage Hexamers. 775

Annu. Rev. Entomol. 36, 205-228. 776

Therneau, T., 2015. A Package for Survival Analysis in S, https://CRAN.R-777

project.org/package=survival. 778

Wallace, D.S., Bairden, K., Duncan, J.L., Eckersall, P.D., Fishwick, G., Holmes, P.H., 779

McKellar, Q.A., Mitchell, S., Murray, M., Parkins, J.J., Stear, M.J., 1999. The influence of 780

increased feeding on the susceptibility of sheep to infection with Haemonchus contortus. 781

Anim. Sci. 69, 457-463. 782

Warner, J.R., 1999. The economics of ribosome biosynthesis in yeast. Trends in biochemical 783

sciences 24, 437-440. 784

White, M.J., Beaver, C.M., Goodier, M.R., Bottomley, C., Nielsen, C.M., Wolf, A.-S.F.M., 785

Boldrin, L., Whitmore, C., Morgan, J., Pearce, D.J., Riley, E.M., 2017. Calorie Restriction 786

Attenuates Terminal Differentiation of Immune Cells. Frontiers in Immunology 7. 787

Whittingham, M.J., Stephens, P.A., Bradbury, R.B., Freckleton, R.P., 2006. Why do we still 788

use stepwise modelling in ecology and behaviour? Journal of Animal Ecology 75, 1182-1189. 789

Wickham, H., 2016. ggplot2: Elegant Graphics for Data Analysis. Springer-Verlag New 790

York. 791

Wiesner, J., Vilcinskas, A., 2010. Antimicrobial peptides The ancient arm of the human 792

immune system. Virulence 1, 440-464. 793

Wilson, K., Cotter, S.C., 2013. Host-Parasite Interactions and the Evolution of Immune 794

Defense, in: Brockmann, H.J., Roper, T.J., Naguib, M., Mitani, J.C., Simmons, L.W., Barrett, 795

L. (Eds.), Advances in the Study of Behavior, Vol 45, pp. 81-174. 796

MANUSCRIP

T

ACCEPTED

ACCEPTED MANUSCRIPT

Wing, E.J., Magee, D.M., Barczynski, L.K., 1988. Acute starvation in mice reduces the 797

number of T-cells and suppresses the development of T-cell-mediated immunity. 798

Immunology 63, 677-682. 799

Woestmann, L., Kvist, J., Saastamoinen, M., 2017. Fight or flight? – Flight increases immune 800

gene expression but does not help to fight an infection. Journal of Evolutionary Biology 30, 801

501-511. 802

Yoshiga, T., Maruta, K., Tojo, S., 1997. Developmental changes of storage proteins and 803

biliverdin-binding proteins in the haemolymph and fat body of the common cutworm, 804

Spodoptera litura. Journal of Insect Physiology 44, 67-76. 805

Zhong, X., Rao, X.-J., Yi, H.-Y., Lin, X.-Y., Huang, X.-H., Yu, X.-Q., 2016. Co-expression 806

of Dorsal and Rel2 negatively regulates antimicrobial peptide expression in the tobacco 807

hornworm Manduca sexta. Scientific reports 6, 20654. 808

Zylberberg, M., 2019. Next-Generation Ecological Immunology*. Physiol. Biochem. Zool. 809

92, 177-188. 810

811

812

MANUSCRIP

T

ACCEPTED

ACCEPTED MANUSCRIPTTables 813

Table 1. Nutritional composition of the 20 chemically-defined diets 814

Diet Protein Carbs Fats Cellulose Micro-nutrients

Energy Ratio P:C

(g/100g diet)

(g/100g diet)

(g/100g diet)

(g/100g diet)

(g/100g diet)

(kJ/100g diet)

(%)

1 10.5 52.5 1.1 33.0 3.0 1112 0.17 1:5

2 7.0 35.0 1.1 54.0 3.0 756 0.17 1:5

3 5.6 28.0 1.1 62.4 3.0 612 0.17 1:5

4 2.8 14.0 1.1 79.2 3.0 326 0.17 1:5

5 21.0 42.0 1.1 33.0 3.0 1112 0.33 1:2

6 14.0 28.0 1.1 54.0 3.0 756 0.33 1:2

7 11.2 22.4 1.1 62.4 3.0 612 0.33 1:2

8 5.6 11.2 1.1 79.2 3.0 326 0.33 1:2

9 31.5 31.5 1.1 33.0 3.0 1112 0.50 1:1

10 21.0 21.0 1.1 54.0 3.0 756 0.50 1:1

11 16.8 16.8 1.1 62.4 3.0 612 0.50 1:1

12 8.4 8.4 1.1 79.2 3.0 326 0.50 1:1

13 42.0 21.0 1.1 33.0 3.0 1112 0.67 2:1

14 28.0 14.0 1.1 54.0 3.0 756 0.67 2:1

15 22.4 11.2 1.1 62.4 3.0 612 0.67 2:1

16 11.2 5.6 1.1 79.2 3.0 326 0.67 2:1

17 52.5 10.5 1.1 33.0 3.0 1112 0.83 5:1

18 35.0 7.0 1.1 54.0 3.0 756 0.83 5:1

19 28.0 5.6 1.1 62.4 3.0 612 0.83 5:1

20 14.0 2.8 1.1 79.2 3.0 326 0.83 5:1

See Table S1 for information about the specific ingredients for each diet 815

816

MANUSCRIP

T

ACCEPTED

ACCEPTED MANUSCRIPT 817

Table 2: Primer and probe sequences used for the qPCR analysis of immune gene expression 818

Gene Primers (5′ - 3′) Probe (5′ - 3′) Amplicon sizes (bp)

Efficiencies

Toll FOR: AATGCTCGTGTTATCATGATCAAA REV: CGTGATCGTAGCCAGCGTTT

VIC-CTGGACCACCACTAACGTCGTCGATTG-TAMRA

76 93.8%

PPO FOR: GCTGTGTTGCCGCAGAATG REV: AAATCCGTGGCGGTGTAGTC

VIC-CCGCGTATCCCGATCATCATCCC-TAMRA

67 97.4%

Lysozyme FOR: TGTGCACAAATGCTGTTGGA REV:CGAACTTGTGACGTTTGTAGATCTTC

VIC-ACATCACCCTAGCTTCTCAGTGCGCC-TAMRA

76 96.6%

Relish FOR: TCAACATAACAACACGGAGGAA REV: ATCAGGTACTAGGCAACTCATATC

6FAM -CCCACAAATTACTTGAAGATGAACAGGACCC-TAMRA

82 95.3%

Moricin FOR: GGCGCAGCGATTGGTAAA REV:GGTTTGAAGAAGGAATAGACATCATG

VIC-TCTCCGGGCGATTAACATAGCCAGC- TAMRA

77 91.4%

EF1 FOR: TCAAGAACATGATCACTGGAACCT REV: CCAGCGGCGACAATGAG

6FAM -CCAGGCCGATTGCGCCGT-TAMRA

94 94.0%

Arylphorin FOR: CGTCAGATGCAGTCTTTAAGATCTTC REV: TGCACGAACCAGTCCAGTTC

VIC-AATACCACGCCAATGGCTATCCGGTT-TAMRA

112 96.7%

Armadillo FOR: TGCACCAGCTGTCCAAGAAG REV: AAAGCGGCAACCATTTGC

6FAM-AAGCTTCTCGCCATGCTATTATGAACTCGC-TAMRA

70 92.8%

Tubulin FOR: CGTGGAGCCCTACAACTCTATCC REV: GCCTCGTTGTCGACCATGA

6FAM-ACCACCCACACCACCCTTGAGCAC-TAMRA

81 100%

819

820

821

MANUSCRIP

T

ACCEPTED

ACCEPTED MANUSCRIPT 822

Table 3: Terms included in each of the basic models describing different attributes of the diet or 823

the amount of protein or carbohydrate consumed. 824

Terms included in model

Model P C P2 C2 Co R Co2 R2 Pe Ce Pe2 Ce2

1 2 X 3 X X 4 X 5 X X 6 X X 7 X X X 8 X X X 9 X X X X 10 X * X 11 X * X X X 12 X 13 X X 14 X 15 X X 16 X X 17 X X X 18 X X X 19 X * X 20 X * X X X 21 X 22 X X 23 X 24 X X 25 X X 26 X X X 27 X X X 28 X X X X 29 X * X 30 X * X X X 825 The table shows the terms included in each of the 30 basic models covering the different dietary 826 attributes. These models were also run including treatment as an additive or interactive effect, giving 827 90 models in total. P (protein) =grams of protein offered, C (carbohydrate) = grams of carbohydrate 828 offered, Co (concentration) = percentage of the diet that comprises digestible nutrients (17%, 34%, 829 42%, 63%), R (ratio) = percentage of protein in the digestible component of the diet (17%, 50% or 830 83%); Pe (protein eaten) =grams of protein eaten, Ce (carbohydrate eaten) = grams of carbohydrate 831 eaten. For Pe and Ce this was calculated over the first 48 hours. Asterisks indicate interactions 832 between terms (e.g. Models 10 and 11 include the interaction between protein and carbohydrate 833 offered). Each of variables was also included as a squared term (e.g. P2). These 30 models were 834 modified by including additive or interactive effects of treatment (base 90 models for all analyses). 835 For the physiological traits and survival, the base 90 models were modified with the additive or 836 interactive effects of expression of the relevant genes (180 models). 837 838

MANUSCRIP

T

ACCEPTED

ACCEPTED MANUSCRIPT 839

Table 4: The best model selected by AICc to explain variation in gene expression across the diet 840

and bacterial treatments. 841

Gene Best Model R2 fixed R2 both

Toll Treat*30 0.059 0.717

PPO Treat*30 0.035 0.715

Lysozyme Treat*30 0.104 0.736

Relish Treat*30 0.120 0.378

Moricin (1) Treat*30 0.741 0.741

Moricin (2) Treat*30 0.097 0.264

Arylphorin Treat*30 0.089 0.275

EF1 Treat*30 0.023 0.862

Armadillo Treat*30 0.034 0.696

Tubulin Treat*30 0.040 0.855

842

Table 5: The best models selected by AICc to explain variation in PPO activity in the 843

haemolymph. GE represents gene expression for the PPO gene. Treat represents the immune 844

challenge treatment. 845

Model df Log Likelihood AICc delta weight R2

3 4 -432.120 872.4 0.00 0.078 0.093

GE+3 5 -431.259 872.7 0.34 0.066 0.098

GE*3 7 -429.321 873.0 0.64 0.057 0.109

Treat+3 5 -431.737 873.7 1.30 0.041 0.095

Treat+GE*3 8 -428.617

873.7 1.34 0.040 0.113

Treat+GE+3 6 -430.716 873.7 1.34 0.040 0.101

7 5 -431.938 874.1 1.70 0.033 0.094

846

MANUSCRIP

T

ACCEPTED

ACCEPTED MANUSCRIPTTable 6: The best models selected by AICc to explain variation in PO activity in the 847

haemolymph. GE represents gene expression for the PPO gene. Treat represents the immune 848

challenge treatment. 849

Model df Log Likelihood AICc delta weight R2

GE*3 7 -425.954 866.3 0.00 0.062 0.092

GE+3 5 -428.058 866.3 0.04 0.061 0.080

3 4 -429.363 866.9 0.58 0.047 0.072

GE+16 5 -428.350 866.9 0.62 0.046 0.078

GE+9 7 -426.378 867.1 0.85 0.041 0.090

16 4 -429.513 867.2 0.88 0.040 0.071

GE+Treat*17 10 -423.239 867.2 0.93 0.035 0.108

Treat*17 9 -424.471 867.5 1.26 0.033 0.100

GE+17 6 -427.775 867.8 1.55 0.029 0.081

GE+10 8 -425.777 868.0 1.75 0.026 0.093

GE+19 6 -427.887 868.0 1.77 0.026 0.081

850

851

Table 7: The best models selected by AICc to explain variation in lytic activity in the 852

haemolymph. GE represents gene expression for the lysozyme gene. 853

Model df Log Likelihood AICc delta weight R2

GE*15 7 647.521 -1280.7 0.00 0.208 0.072

GE*18 9 649.465 -1280.3 0.34 0.176 0.080

GE*3 5 644.526 -1278.9 1.82 0.084 0.051

854

855

856

857

MANUSCRIP

T

ACCEPTED

ACCEPTED MANUSCRIPTTable 8: The best models selected by AICc to explain variation in survival after bacterial (dead 858

or live) injection. GE represents gene expression for the Moricin gene. 859

Model df Log Likelihood AICc delta weight R2

Treat+GE*30 12 -1662.917

3350.8 0.00 0.139 0.127

Treat*GE*30 23 -1650.823

3351.1

0.30 0.120 0.186

Treat+GE+30 7 -1668.715

3351.8

0.99 0.085 0.098

Treat+GE+20 7 -1668.864

3352.1

1.29 0.073 0.097

GE*30 11 -1664.796

3352.4

1.61 0.062 0.118

860

861

862

863

MANUSCRIP

T

ACCEPTED

ACCEPTED MANUSCRIPT

Figure legends 864

Figure 1 – The total amount of food eaten by caterpillars that were either (a) handled or (b) 865

injected with dead bacteria (Experiment 1) or (d) injected with dead bacteria versus (e) live 866

bacteria (Experiment 2). Blue colours indicate low consumption and red colours high 867

consumption. Colours are standardised within each experiment and are not comparable across 868

experiments. Numbers on the contour lines indicate z values for consumption. Intake arrays 869

indicating total consumption across the diets are shown separately for (c) experiment 1 and (d) 870

experiment 2. 871

872

873

874

MANUSCRIP

T

ACCEPTED

ACCEPTED MANUSCRIPT 875

Figure 2 – Mean gene expression (+/- SE) for each of the immune genes and non-immune genes 876

in response to immune challenge treatment, relative to the ‘handled’ controls. Genes are 877

grouped by immune pathway Toll (blue zone: Toll, PPO, Lysozyme and Moricin), IMD (pink 878

zone: Moricin and Relish [11] [12] ) or classified as non-immune genes (grey zone; Arylophorin, 879

EF1, Armadillo and Tubulin). The black dashed line represents gene expression in the handled 880

group. Residuals from the model accounting for the random effects of ‘plate’ are plotted against 881

treatment. All models were re-run on untransformed data for ease of visualisation. 882

883

884

MANUSCRIP

T

ACCEPTED

ACCEPTED MANUSCRIPTFigure 3 – Variation in Moricin expression across diets in haemolymph of caterpillars subject to 885

different immune challenge treatments, (a) handled only, (b) injected with dead bacteria (Expt 886

1), (c) injected with dead bacteria (Expt 2) and (d) injected with live bacteria. Blue colours 887

indicate low gene expression and red colours high gene expression. Colours are standardised 888

within each experiment and are not comparable across experiments. 889

890

891

892

MANUSCRIP

T

ACCEPTED

ACCEPTED MANUSCRIPT 893

Figure 4 – Variation in immune gene expression across diets in haemolymph of caterpillars 894

subject to different immune challenge treatments, (a-c) Toll, (d-f) PPO, (g-i) Lysozyme and (j-l) 895

Relish. All figures in the first column are for the handled treatment, column 2 includes those 896

injected with dead bacteria and column 3, those injected with live bacteria. Blue colours 897

indicate low gene expression and red colours high gene expression. 898

899

900

901

902

MANUSCRIP

T

ACCEPTED

ACCEPTED MANUSCRIPT 903

Figure 5 – Variation in non-immune gene expression across diets in haemolymph of caterpillars 904

subject to different immune challenge treatments, (a-c) Arylophorin, (d-f) EF1, (g-i) Armadillo 905

and (j-l) Tubulin. All figures in the first column are for the handled treatment, column 2 906

includes those injected with dead bacteria and column 3, those injected with live bacteria. Blue 907

colours indicate low gene expression and red colours high gene expression. 908

909

910

911

MANUSCRIP

T

ACCEPTED

ACCEPTED MANUSCRIPTFigure 6 – Physiological immune responses vary with the protein content of the diet and the 912

expression of the relevant gene. (a) PPO and (b) PO activity in the haemolymph in response to 913

PPO gene expression and (c) lysozyme activity in the haemolymph in response to lysozyme gene 914

expression. Blue colours indicate low activity and red colours high activity. 915

916

917

918

919

MANUSCRIP

T

ACCEPTED

ACCEPTED MANUSCRIPTFigure 7 –Survival for larvae injected with either dead (a,c) or live (b,d) X. nematophila 920

bacteria. Predicted survival curves (a,b) are plotted for model Treat*GE*30. Protein eaten and 921

carbohydrate eaten were set at mean values for each coefficient and Moricin gene expression 922

was set as either low (lower quartile) or high (upper quartile). To visualise the effects of diet on 923

survival, time to death (c,d) is plotted as thin plate splines against the amount of protein and 924

carbohydrate consumed. Blue colours indicate a short time to death and red colours a slow time 925

to death. 926

927

928

MANUSCRIP

T

ACCEPTED

ACCEPTED MANUSCRIPT 929

MANUSCRIP

T

ACCEPTED

ACCEPTED MANUSCRIPT

Highlights

• High protein diets improved survival after live bacterial infection

• Injection with dead bacteria increased expression of Toll and IMD immune genes.

• Injection with live bacteria inhibited immune gene expression (GE).

• The ratio and concentration of macronutrients in the diet affected GE.

• GE only predicted functional immune activity at high levels of dietary protein.